TVnet: automated time-resolved tracking of the tricuspid valve plane in MRI long-axis cine images with a dual-stage deep learning pipeline
Tracking the tricuspid valve (TV) in magnetic resonance imaging (MRI) long-axis cine images has the potential to aid in the evaluation of right ventricular dysfunction, which is common in congenital heart disease and pulmonary hypertension. However, this annotation task remains difficult and time-de...
| Main Authors: | , , , , , |
|---|---|
| Formato: | Conference item |
| Idioma: | English |
| Publicado em: |
Springer
2021
|
| _version_ | 1826312155223293952 |
|---|---|
| author | Gonzales, RA Lamy, J Seemann, F Heiberg, E Onofrey, JA Peters, DC |
| author_facet | Gonzales, RA Lamy, J Seemann, F Heiberg, E Onofrey, JA Peters, DC |
| author_sort | Gonzales, RA |
| collection | OXFORD |
| description | Tracking the tricuspid valve (TV) in magnetic resonance imaging (MRI) long-axis cine images has the potential to aid in the evaluation of right ventricular dysfunction, which is common in congenital heart disease and pulmonary hypertension. However, this annotation task remains difficult and time-demanding as the TV moves rapidly and is barely distinguishable from the myocardium. This study presents TVnet, a novel dual-stage deep learning pipeline based on ResNet-50 and automated image linear transformation, able to automatically derive tricuspid annular plane systolic excursion. Stage 1 uses a trained network for a coarse detection of the TV points, which are used by stage 2 to reorient the cine into a standardized size, cropping, resolution, and heart orientation and to accurately locate the TV points with another trained network. The model was trained and evaluated on 4170 images from 140 patients with diverse cardiovascular pathologies. A baseline model without standardization achieved a Euclidean distance error of 4.0 ± 3.1 mm and a clinical-metric agreement of ICC = 0.87, whereas a standardized model improved the agreement to 2.4 ± 1.7 mm and an ICC = 0.94, on par with an evaluated inter-observer variability of 2.9 ± 2.9 mm and an ICC = 0.92, respectively. This novel dual-stage deep learning pipeline substantially improved the annotation accuracy compared to a baseline model, paving the way towards reliable right ventricular dysfunction assessment with MRI. |
| first_indexed | 2024-03-07T08:23:21Z |
| format | Conference item |
| id | oxford-uuid:83c83c06-745d-40e6-ae7f-e5c61d5abe46 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T08:23:21Z |
| publishDate | 2021 |
| publisher | Springer |
| record_format | dspace |
| spelling | oxford-uuid:83c83c06-745d-40e6-ae7f-e5c61d5abe462024-02-09T14:37:47ZTVnet: automated time-resolved tracking of the tricuspid valve plane in MRI long-axis cine images with a dual-stage deep learning pipelineConference itemhttp://purl.org/coar/resource_type/c_5794uuid:83c83c06-745d-40e6-ae7f-e5c61d5abe46EnglishSymplectic ElementsSpringer2021Gonzales, RALamy, JSeemann, FHeiberg, EOnofrey, JAPeters, DCTracking the tricuspid valve (TV) in magnetic resonance imaging (MRI) long-axis cine images has the potential to aid in the evaluation of right ventricular dysfunction, which is common in congenital heart disease and pulmonary hypertension. However, this annotation task remains difficult and time-demanding as the TV moves rapidly and is barely distinguishable from the myocardium. This study presents TVnet, a novel dual-stage deep learning pipeline based on ResNet-50 and automated image linear transformation, able to automatically derive tricuspid annular plane systolic excursion. Stage 1 uses a trained network for a coarse detection of the TV points, which are used by stage 2 to reorient the cine into a standardized size, cropping, resolution, and heart orientation and to accurately locate the TV points with another trained network. The model was trained and evaluated on 4170 images from 140 patients with diverse cardiovascular pathologies. A baseline model without standardization achieved a Euclidean distance error of 4.0 ± 3.1 mm and a clinical-metric agreement of ICC = 0.87, whereas a standardized model improved the agreement to 2.4 ± 1.7 mm and an ICC = 0.94, on par with an evaluated inter-observer variability of 2.9 ± 2.9 mm and an ICC = 0.92, respectively. This novel dual-stage deep learning pipeline substantially improved the annotation accuracy compared to a baseline model, paving the way towards reliable right ventricular dysfunction assessment with MRI. |
| spellingShingle | Gonzales, RA Lamy, J Seemann, F Heiberg, E Onofrey, JA Peters, DC TVnet: automated time-resolved tracking of the tricuspid valve plane in MRI long-axis cine images with a dual-stage deep learning pipeline |
| title | TVnet: automated time-resolved tracking of the tricuspid valve plane in MRI long-axis cine images with a dual-stage deep learning pipeline |
| title_full | TVnet: automated time-resolved tracking of the tricuspid valve plane in MRI long-axis cine images with a dual-stage deep learning pipeline |
| title_fullStr | TVnet: automated time-resolved tracking of the tricuspid valve plane in MRI long-axis cine images with a dual-stage deep learning pipeline |
| title_full_unstemmed | TVnet: automated time-resolved tracking of the tricuspid valve plane in MRI long-axis cine images with a dual-stage deep learning pipeline |
| title_short | TVnet: automated time-resolved tracking of the tricuspid valve plane in MRI long-axis cine images with a dual-stage deep learning pipeline |
| title_sort | tvnet automated time resolved tracking of the tricuspid valve plane in mri long axis cine images with a dual stage deep learning pipeline |
| work_keys_str_mv | AT gonzalesra tvnetautomatedtimeresolvedtrackingofthetricuspidvalveplaneinmrilongaxiscineimageswithadualstagedeeplearningpipeline AT lamyj tvnetautomatedtimeresolvedtrackingofthetricuspidvalveplaneinmrilongaxiscineimageswithadualstagedeeplearningpipeline AT seemannf tvnetautomatedtimeresolvedtrackingofthetricuspidvalveplaneinmrilongaxiscineimageswithadualstagedeeplearningpipeline AT heiberge tvnetautomatedtimeresolvedtrackingofthetricuspidvalveplaneinmrilongaxiscineimageswithadualstagedeeplearningpipeline AT onofreyja tvnetautomatedtimeresolvedtrackingofthetricuspidvalveplaneinmrilongaxiscineimageswithadualstagedeeplearningpipeline AT petersdc tvnetautomatedtimeresolvedtrackingofthetricuspidvalveplaneinmrilongaxiscineimageswithadualstagedeeplearningpipeline |